After four months of steady development and 578 commits, I’m happy to share Portfolio OS v1.0.0 — a personal experiment that turned into a fully functional, self-documenting development platform. What started as a simple portfolio site became a deep dive into automation, documentation, and AI-assisted workflows that I now use in my everyday development routine.
🔗 Live Site: johnschibelli.dev (retired) 📖 Documentation: docs.johnschibelli.dev (retired) 💻 GitHub: github.com/jschibelli/portfolio-os
If you’d like to see how the system works in practice, check out the accompanying case study — it walks through the architecture, automation, and design thinking behind the project.
The Journey: From Portfolio to Platform
When I began this project in July 2025, my only goal was to build a clean, modern portfolio using Next.js 15 and TypeScript. But as I worked, I kept running into the same issues developers face — repetitive tasks, outdated docs, manual testing — and it got me thinking: what if the portfolio itself could solve those problems?
So, instead of just showcasing my work, I built something that does the work.
The Core Idea: Self-Documenting + AI-Assisted Development
Portfolio OS revolves around three ideas that shaped the way I work:
1. Automation Over Manual Labor
I built more than 100 PowerShell scripts that handle everything from pull request creation and changelog generation to issue management and release prep. They’re not for show — I use them daily, and they’ve saved countless hours of repetitive work.
2. Self-Documenting by Design
The platform updates its own documentation when the code changes. Changelogs, examples, and API references all stay in sync. It’s not about perfection — it’s about eliminating drift so that what’s written always reflects what’s running.
3. Multi-Agent AI Workflows
I set up two AI agents (Jason and Chris) that help with development tasks in isolated worktrees. They run checks, manage PRs, and handle smaller jobs in parallel. It’s not about replacing developers — it’s about amplifying what one person can do with the right systems in place.
The Stack Behind It
Frontend: Next.js 15, React 18, TypeScript, Tailwind CSS, Radix UI, Framer Motion Backend: Prisma ORM, SQLite (dev) / PostgreSQL (prod), Hashnode GraphQL API, Vercel Blob DevOps: Turborepo, GitHub Actions, Playwright, Jest, PowerShell automation AI: GPT-4 integration for content generation, multi-agent workflows, automated PR management
What It Can Do
- Dynamic Case Studies and blog content via Hashnode API
- Self-Updating Documentation tied to commits and releases
- Automation Layer that manages PRs, issues, and changelogs
- Integrated Admin Dashboard for writing, media management, and analytics
- Full Testing Pipeline with Playwright and Jest for quality control
The Impact
Over four months, automation became a silent teammate. PRs now open and merge automatically. Documentation stays up to date. Tests run without me remembering to run them. The payoff isn’t flashy — it’s peace of mind. The platform just works.
More importantly, the process taught me a lot:
- Investing in automation early compounds over time.
- Type safety and testing prevent silent regressions.
- Documentation isn’t an afterthought — it’s a byproduct of good systems.
- AI is most effective when treated as a collaborator, not a replacement.
Lessons Learned
- Automate what you repeat. Anything done twice deserves a script.
- Keep docs close to code. If they live together, they’ll stay together.
- AI can help, but it still needs direction. It’s great at execution, not judgment.
- Small releases build big progress. Monthly versioning kept me accountable.
- Measure performance and accessibility early. Fixing them later costs more.
What’s Next (v1.1.0 Roadmap)
I’m already sketching ideas for the next update:
- A more detailed analytics dashboard
- Smarter AI agent collaboration using vector search
- Optional team collaboration mode
- Additional integrations (GitHub, Notion, Linear)
- Continued performance optimization and accessibility improvements
Explore the Codebase
BASH
# Clone the repo
git clone https://github.com/jschibelli/portfolio-os.git
cd portfolio-os
# Install dependencies
pnpm install
# Start dev server
pnpm dev
Docs and a walkthrough were originally published at docs.johnschibelli.dev (retired). For a more detailed breakdown, the Portfolio OS Case Study covers the full development process and automation flow.
Final Thoughts
This project isn’t about perfection or bragging rights — it’s about progress. I wanted a platform that could evolve alongside me, where every improvement teaches something new. Building Portfolio OS reminded me that automation isn’t just about saving time; it’s about creating space to think, experiment, and grow.
If you’re curious about how it all works under the hood, I encourage you to explore the case study or browse the GitHub repo.
Built with care using Next.js, TypeScript, and AI-assisted workflows.